Accéder directement au contenu Accéder directement à la navigation
Communication dans un congrès

Data-Flow/Dependence Profiling for Structured Transformations

Fabian Gruber 1, 2, 3, 4 Manuel Selva 1, 2, 3, 4 Diogo Sampaio 1, 2, 3, 4 Christophe Guillon 5 Antoine Moynault 5 Louis-Noël Pouchet 6 Fabrice Rastello 1, 2, 3, 4
1 UGA [2016-2019] - Université Grenoble Alpes [2016-2019]
2 CORSE - Compiler Optimization and Run-time Systems
Inria Grenoble - Rhône-Alpes, LIG - Laboratoire d'Informatique de Grenoble
3 Grenoble INP - Institut polytechnique de Grenoble - Grenoble Institute of Technology
4 Laboratoire d'Informatique de Grenoble
5 STMicroelectronics
6 CSU - Colorado State University [Fort Collins]
Abstract : Profiling feedback is an important technique used by developers for performance debugging, where it is usually used to pinpoint performance bottlenecks and also to find optimization opportunities. Assessing the validity and potential benefit of a program transformation requires accurate knowledge of the data flow and dependencies, which can be uncovered by profiling a particular execution of the program. In this work we develop poly-prof, an end-to-end infrastructure for dynamic binary analysis, which produces feedback about the potential to apply complex program rescheduling. Our tool can handle both inter-and intraproce-dural aspects of the program in a unified way, thus providing interprocedural transformation feedback.
Keywords : Performance Feedback Polyhedral Model Loop Transformations Compiler Optimization Binary Instrumentation Dynamic Dependence Graph
Type de document :
Communication dans un congrès
Liste complète des métadonnées

Littérature citée [77 références]  Voir  Masquer  Télécharger
https://hal.inria.fr/hal-02060796
Contributeur : Fabian Gruber <>
Soumis le : jeudi 7 mars 2019 - 16:15:34
Dernière modification le : jeudi 19 novembre 2020 - 13:02:08
Archivage à long terme le : : dimanche 9 juin 2019 - 10:28:06

Fichier

Vignette du document
main.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

CNRS | INRIA | LIG | LIG-SRCPR-CORSE | LIG_SRCPR | UGA

Citation

Fabian Gruber, Manuel Selva, Diogo Sampaio, Christophe Guillon, Antoine Moynault, et al.. Data-Flow/Dependence Profiling for Structured Transformations. PPoPP 2019 - 24th Symposium on Principles and Practice of Parallel Programming, Feb 2019, Washington, D.C., United States. pp.173-185, ⟨10.1145/3293883.3295737⟩. ⟨hal-02060796⟩

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

248

Téléchargements de fichiers

1100